Did it finally happen?

 In recent years, a lot of people have said that there's been a decline in social connections, resulting in an "epidemic of loneliness."  I've had a number of posts arguing that there's no evidence in favor of this idea, and some evidence against it.  But recently, a survey by the Survey Center on American Life repeated a question used by the Gallup Poll and found a substantial decline in the number of close friends that people say they have.  The question is "Not counting your relatives, about how many close friends would you say you have?"  Gallup asked this a number of times from 1976 to 2003 and always found about 2-3% answering zero--in 2021, that jumped to 12%.   Moreover, the number of people reporting a large number of close friends dropped sharply.  A detailed comparison between 2003 and 2021:

            2003          2021

0            2%        12% 

1            6%        7%

2            8%       13%

3            11%      17%

4-5         28%      24%

6-9        18%        12%

10+        27%       13%            

 That's a big difference, and the surveys used the same question.  However, there were a number of differences between them that might be relevant.  The Survey Center on American Life notes that its survey was done online, while the Gallup surveys were done over the phone, and says "respondents talking to a live interviewer might have been less willing to report having no friends or inflate the number of close friends they had out of concerns about how their responses would be perceived by the interviewer."  There are also a couple of other differences.  The Gallup surveys didn't have a specific focus, but contained a miscellaneous selection of questions.  The 2021 survey, in contrast, focused on friendship, and the question about number of friends was preceded by a series of questions on how often you did various things with your friends (including not only things like texting or e-mailing, but "had a private conversation in which you shared personal feelings or problems" and "told a friend that you loved them") and the kinds of people you were friends with.  Also, the question on the number of close friends was introduced with "Now, thinking only about the friends you are close to..."  I think that the combination of those features almost certainly pushed people towards being more selective in their definition of "close friends" than they were in the Gallup surveys.  Moreover, the series of general questions about friends might have led some people to expect that they would be asked follow-up questions about specific friends ("Now thinking of the person you consider your best friend, when was the last time you...).  So some people may have given lower reports of the number of friend because they didn't want the survey to go on too long.  

This is not to say that the Gallup surveys are better, just that they are different in ways that might account for the differences in the number of friends that people reported.  So although the numbers in 2021 are very different from the numbers in 2003, I still don't think that there's any convincing evidence of a decline in friendship.

[Data from the Roper Center for Public Opinion Research]